From honing TV shows into long-running hits, to getting more accurate
feedback from consumer panels, key pads are proving a very effective
research tool, says Ken Gofton
Fourteen months ago, on board the Canberra, Howell Henry Chaldecott
Lury’s Adam Lury spoke on interactivity. To demonstrate his point, he
came with notes and back-up material on a long list of related topics.
The audience was armed with key pads and invited to vote on which
subject they wanted to hear first.
But Lury went further. In a brave, even foolhardy gesture, he asked the
audience to use the key pads to indicate when they were bored. Once 15%
had lost interest in a topic, everyone voted on what subject he should
turn to next.
Key pads, linked to a computer, are being used increasingly at
conferences to provide an instant analysis of delegates’ opinions and
Similar technology is being used by market researchers in other fields,
especially to test new concepts. This article discusses two techniques
which originated in North America and are now being used in the UK.
Ready Steady Cook
The public doesn’t think a lot about how new programmes arrive on
screen. Presumably, it thinks, someone submits an idea which gets kicked
around by ‘them upstairs’ and then wins approval and a budget, or dies.
Some ideas are obvious winners - like Pride and Prejudice and Cracker -
but others must be absolute dogs which are quietly put to sleep. In
between are those programmes with promise and which can be improved if
their weaknesses are correctly identified and eliminated. Here, as in
other forms of new product development, market research techniques have
a pivotal role.
According to Nino Cirone, director of the Pegram Walters’ subsidiary
Broadcast Research, the situation was a little different with BBC2’s hit
Ready, Steady, Cook. The programme was developed by Bazal Productions,
the producers of Food and Drink, to meet a strategic brief. What was
needed was an alternative to BBC1’s children’s programmes for the late
afternoon audience of mainly older women.
‘In this case, the idea was clearly a good one,’ says Cirone. ‘The
research role was to see if there were ways of tweaking it further.
Ready, Steady, Cook had to hit the ground running, because it was up
against strong competition from quiz shows on Channel 4.’
Ready Steady Cook is half game show and half cookery programme. Two
chefs pit their skills against each other, creating and cooking a dish
in 20 minutes from ingredients they haven’t seen in advance. The
ingredients are provided by two members of the audience, who then act as
the chef’s assistants.
With a pilot edition in the can, Broadcast Research was called in to
gauge its appeal, assess the public’s ‘propensity to view’ and provide
feedback on elements within the programme itself.
The technique, called PEAC, originated in Canada and has been used by
Broadcast Research for more than 50 programming studies in the last five
years. The method is also suitable for testing advertising or
promotions. A sample group watches the programme, each with a hand-held
key pad to relay their reactions to a computer via a radio link. There
are three key elements:
* Moment-by-moment rating of the material - respondents are asked to key
in their reactions throughout a programme on a five-point scale, from
very interesting/enjoyable down to not-at-all interesting/enjoyable. The
results can later be overlaid on the programme as a simple graph
* Pre-coded questions, such as: ‘Is this a good idea for a series’, ‘Is
it for people like me’, ‘Is it easy to follow’, ‘Is it fun?’
* Qualitative discussion with a sample of the respondents, to probe into
why some elements of the programme are liked or disliked, or why the
pace might flag at some point.
Two research sessions were held, one in Yorkshire and one in London,
with 25 viewers at each. Ready, Steady, Cook came through with flying
colours, with the research confirming it as enjoyable and entertaining.
The key lay in its mix of cookery and competition.
Most respondents interpreted it as a cookery programme, with the game
show elements and time pressures adding excitement and the banter
between the contestants passing information on in a friendly way. To the
surprise of the producers, what the viewers liked most about the recipes
were the professional tips on enhancing and presenting the dishes.
But three elements in the pilot programme got the thumbs down. Repeated
chanting by the audience of the catch phrase ‘ready, steady, cook’ was
dismissed as ‘embarrassing’. And the original idea for ending the show,
with the contestants guessing the number of pickled onions in a jar or
the weight of a marrow, seemed to have no link with the rest of the
show. It was the respondents who suggested the ending that was finally
adopted, of having the audience vote which dish was the winner.
Finally, the prizes on offer to the studio participants were criticised.
In the pilot programme, the winner got her last grocery bill paid, which
was seen as ‘nothing special’, while the loser got an ‘insulting’ wooden
spoon. Now the programme’s on air, the winner gets a pounds 100 cheque,
the loser a luxury hamper.
Nine out of ten women and two out of three men thought the programme was
enjoyable and 70% of the total sample thought it was a good idea for a
On these figures, a trial run of 24 programmes, eventually extended to
94, was commissioned to start last autumn. During its run, it became the
most popular daytime non-news programme with housewives. Screening was
increased from three to five times a week and audiences intermittently
averaged over 3 million, against an original target of 1.5 million.
‘It’s a nerve-wracking experience for a producer to see a graph
recording an audience’s second-by-second reactions,’ says Peter
Bazalgette, managing director of Bazal Productions.
‘We knew that the last five minutes were a bit of a mess and the
research helped us sort it out. It enabled us to get the prize structure
acceptable to the viewers and to achieve the right tone - the pilot was
‘Mass concept testing’
Marketers know that new product development is the lifeblood of the
business. They also know how the odds are stacked against new products
making the grade.
A team which broke away from Procter & Gamble in the US to form an NPD
consultancy decided that the way to get more winning products was to
start with a lot more concepts. But those ideas would need screening and
- being P&G trained - the team wanted to back their decisions with
figures, not emotional judgements.
A dollars 1m development programme led to a patented system called
AcuPOLL. After running in the US for four years, it has a growing client
list that includes P&G, Johnson & Johnson, Van den Bergh, Colgate
Palmolive, Disney, Pepsi and AT&T. It’s been available in Europe since
April through London-based Q1 International Research. The first study on
this side of the Atlantic was for Colgate - in France, Italy and
Although the main application is different, there are similarities
between AcuPOLL and Broadcast Research’s PEAC. An AcuPOLL session
involves 100-150 consumers, with each respondent given a computer-linked
key pad with numbers running from nought to ten. Over three hours, they
can examine up to 40 concepts and answer 300-400 pre-set questions -
some of them demographic or psychographic - and as the answers are
tapped in, the computer provides instant analysis. At the end, there’s
an opportunity for the moderator to probe more deeply into why some of
the answers were given, which is a key part of the process. It is the
sheer numbers involved that startle some UK researchers. ‘The system is
different and novel and breaks some research rules,’ admits Q1 managing
director Ray Gundersen.
The main area of concern is over what researchers call the ‘order
effect’. Put simply, the more questions interviewees are asked, the more
tired and bored they get, so questions asked towards the end of a
session are likely to produce lower scores than those asked near the
beginning. The conventional solution is to have more frequent, smaller
sessions and to rotate the questions, but with up to 400 questions being
posed, that’s not practicable.
In the early days, AcuPOLL did suffer from ‘order effect’. This has been
overcome, Gundersen claims, by eliminating verbiage from the questions,
by regularly breaking up the sessions with less serious questions - on
lifestyles and so on - and in particular, by modifying the software to
compensate. He believes Q1 can demonstrate that order effect has been
reduced to the point where it is not statistically significant.
Client companies can book full or half sessions exclusive to themselves,
but Q1 also runs omnibus sessions which a number of companies can share
- monthly with panels of housewives, quarterly with teenagers and male
drinkers. The omnibus sessions began in the UK and are now being
extended to the major Continental countries.
The computer produces a mean score for each question and also a simpler
A, B, C or D rating. It is able to compare the scores of all the
concepts being tested on the day, or make comparisons with others in a
database stretching back 18 months. Currently, US tests dominate the
database, but the number of European examples is growing fast.
‘International comparisons are very interesting,’ says Gundersen. ‘You
find the Italians are very extreme - they really like an idea, or they
really reject it. The Germans are more pragmatic. The French and the
Americans are more like the Italians, the British more like the Germans.
‘But when you allow for the different levels of scoring, a good idea is
usually a good idea across all countries. The exceptions are where there
are cultural differences. French housewives love cleaning products
containing bleach, for example, German housewives don’t. And the English
still don’t like the idea of iced tea.’